Globally, tobacco use accounts for 4.9 million deaths each year. These deaths are caused by inhaled smoke toxicants. Extensive research describes cigarette smoke toxicant content (or yield), a vital link to understanding user toxicant exposure. However, millions of people use waterpipes (WP) to smoke tobacco. WP tobacco smoking may also be deadly, though little is known about WP smoke toxicant yield. A WP (hookah, shisha) has a hose, water bowl, body, and a head filled with sweetened and flavored tobacco that is heated with charcoal. WP tobacco smoking is increasing in the U.S., especially among college students. Relative to one cigarette, one WP use episode can generate 100 times the smoke. Current technology demands that, if we are to learn about WP smoke, we must use machines to generate it. Machine-generated smoke may limit validity because generation of WP smoke involves several key variables that differ by user and use episode which cannot be standardized in the analytical lab. This response to PAR-08-212 integrates mechanical engineering, chemistry, and clinical behavioral pharmacology to develop and test an unobtrusive WP smoke sampling instrument for the purpose of assessing the toxicant content of smoke produced in the natural environment. The specific aims are to: 1) Characterize a real-time WP smoke sampling device (Years 1-3). We have developed a prototype device that measures puff topography and samples WP smoke in real- time (REALTIME). Because particle deposition in the device may alter smoke toxicant content, correction factors need to be developed to infer true toxicant content from that measured. We will therefore measure toxicant-specific deposition fractions in the device, in particular for: particle number, particle mass, nicotine, tobacco-specific nitrosamines, and polycyclic aromatic hydrocarbons. 2) Evaluate REALTIME design options in the clinical laboratory (Years 1 and 2). Two design options, placement of the smoke sampler and the percentage of smoke sampled, involve tradeoffs between accurate measurement of toxicant content and realistic smoker behavior. We propose two controlled clinical laboratory studies to inform design option choice, with Study 1 addressing sampler placement (bowl or mouthpiece) and Study 2 addressing percentage of smoke sampled (2, 5, or 10%). The preferred option maximizes realistic smoke toxicant analysis while minimizing effects on smoker behavior. 3) Evaluate REALTIME in the natural environment (years 4 and 5). REALTIME will be attached to WPs used by consenting individual and groups of WP cafi patrons. The study will demonstrate REALTIME's use in natural settings and examine the relationship between the number of WP users and tobacco temperature and WP smoke toxicant yield. In sum, the technology and methodology developed in this project will enable valid analysis of WP smoke toxicant content. This interdisciplinary project addresses an emerging public health threat and is directly relevant to the NIH mission and wholly responsive to PAR-08-212.